上海交通大学学报(英文版) ›› 2013, Vol. 18 ›› Issue (3): 366-375.doi: 10.1007/s12204-013-1407-0

• 论文 • 上一篇    下一篇

Online Sequential Extreme Learning Machine Based Multilayer Perception with Output Self Feedback for Time Series Prediction

PAN Feng1* (潘 峰), ZHAO Hai-bo2 (赵海波)   

  1. (1. School of Information Science and Technology, Donghua University, Shanghai 200051, China; 2. Institude of Aircraft Equipment, Naval Academy of Armament, Shanghai 200436, China)
  • 出版日期:2013-06-28 发布日期:2013-08-12
  • 通讯作者: PAN Feng(潘 峰) E-mail:fpan@dhu.edu.cn

Online Sequential Extreme Learning Machine Based Multilayer Perception with Output Self Feedback for Time Series Prediction

PAN Feng1* (潘 峰), ZHAO Hai-bo2 (赵海波)   

  1. (1. School of Information Science and Technology, Donghua University, Shanghai 200051, China; 2. Institude of Aircraft Equipment, Naval Academy of Armament, Shanghai 200436, China)
  • Online:2013-06-28 Published:2013-08-12
  • Contact: PAN Feng(潘 峰) E-mail:fpan@dhu.edu.cn

摘要: This study presents a time series prediction model with output self feedback which is implemented based on online sequential extreme learning machine. The output variables derived from multilayer perception can feedback to the network input layer to create a temporal relation between the current node inputs and the lagged node outputs while overcoming the limitation of memory which is a vital part for any time-series prediction application. The model can overcome the static prediction problem with most time series prediction models and can effectively cope with the dynamic properties of time series data. A linear and a nonlinear forecasting algorithms based on online extreme learning machine are proposed to implement the output feedback forecasting model. They are both recursive estimator and have two distinct phases: Predict and Update. The proposed model was tested against different kinds of time series data and the results indicate that the model outperforms the original static model without feedback.

关键词: time series prediction, extreme learning machine (ELM), autoregression (AR), online sequential learning ELM (OS-ELM), recurrent neural network (RNN)

Abstract: This study presents a time series prediction model with output self feedback which is implemented based on online sequential extreme learning machine. The output variables derived from multilayer perception can feedback to the network input layer to create a temporal relation between the current node inputs and the lagged node outputs while overcoming the limitation of memory which is a vital part for any time-series prediction application. The model can overcome the static prediction problem with most time series prediction models and can effectively cope with the dynamic properties of time series data. A linear and a nonlinear forecasting algorithms based on online extreme learning machine are proposed to implement the output feedback forecasting model. They are both recursive estimator and have two distinct phases: Predict and Update. The proposed model was tested against different kinds of time series data and the results indicate that the model outperforms the original static model without feedback.

Key words: time series prediction, extreme learning machine (ELM), autoregression (AR), online sequential learning ELM (OS-ELM), recurrent neural network (RNN)

中图分类号: